The long term goal for this core is to provide highly accurate medical sequencing data in a high throughput pipeline at the lowest possible cost to the PPG. The significance of genetic changes identified by this effort will be evaluated in Project 2 of the PPG, and by investigators from around the world. Specific Aim: Using high-throughput, directed PCR and DMA sequencing, we will define the mutations and polymorphisms present in candidate genes using DNA samples derived from a cohort of 94 AML patients for which we have banked both tumor (bone marrow) and germline (skin) samples. For over two years, a production pipeline for directed PCR and sequencing of selected regions of individual human genomes has been in operation at the Genome Sequencing Center (GSC) at Washington University. We have developed an automated primer selection system to characterize the optimal tiling path of amplicons across any region of interest in the genome. The bar-coding of both samples and amplification primers preserves integrity as each sample moves through the production pipeline. Any combination of sample and amplicon is permitted due to the flexibility that we have built into the sample storage systems, automated platforms, and Oracle LIMS databases. Since all amplification primers have universal (forward and reverse) tails, all sequencing reactions are performed with the same parameters. From 5/06 to 5/07, we have completed -258,000 reads with a sequencing pass rate of over 90%. We also have developed an automated alternative chemistry pipeline for amplicons, which are identified during primer selection as difficult-to-sequence. We analyze the resulting sequence data with two independent, automated systems that identify mutations/polymorphisms, insertions and deletions. Rapid manual review of the automated pipeline analysis reports and sequence data, when necessary, refines and clarifies the automated output. For the PPG effort, we will produce over 15,000 sequencing reads/month. The initial re-sequencing will be performed on 94 AML tumor samples. When sequence changes that are predicted to alter gene function are detected in the tumor samples, we will sequence the same amplicons in the matching germline set to determine whether the change is somatic, or also present in the germline. Once a sequence change is confirmed to be non-synonymous and somatic, the same amplicon will be sequenced using material from the 94 CALGB "validation" samples to further assess overall mutational frequency. All of these components taken together comprise a high-throughput system that produces high-quality medical sequencing information at the lowest possible cost.